1. Statement of the Technical Field
The invention concerns computing systems. More particularly, the invention concerns computing systems and methods for efficient comparative non-spatial image data analysis.
2. Description of the Related Art
Biometric systems are often used to identify individuals based on their unique traits in many applications. Such applications include security applications and forensic applications. During operation, the biometric systems collect biometric image data defining images comprising visual representation of physical biometric markers. The physical biometric markers include facial features, fingerprints, hand geometries, irises and retinas. Physical biometric markers are present in most individuals, unique to the individuals, and permanent throughout the lifespan of the individuals.
The identity of a person can be determined by an expert technician using an Automatic Biometric Identification System (“ABIS”). The ABIS generally compares a plurality of biometric images to a reference biometric image to determine the degree of match between content thereof. Thereafter, the ABIS computes a match score for each of the plurality of biometric images. Biometric images with match scores equal to or greater than a threshold value are selected as candidate biometric images. The expert technician then reviews the candidate biometric images for purposes of identifying the individual comprising the physical biometric marker visually represented within the reference biometric image. This expert review typically involves a manual one-to-one examination of the reference biometric image to each of the candidate biometric images. Often times, there are a relatively large number of candidate biometric images that need to be reviewed by the expert technician. As such, the identification process may take a considerable amount of time and resources to complete.